AI Computer Vision Engineer

Software Engineer, Computer Vision and Deep Learning

Developing new computer vision algorithms with founders in C/C++ and Python for solving challenging real-world problems, coming up with large scale data collection techniques for training Deep Neural Nets, driving the development of new algorithms that dramatically improve existing methods, researching and maintaining state-of-the-art ML/CV algorithms that can analyze images, and coding full-stack building products from end to end.

Lead the research, development, and deployment of state-of-the-art deep learning models for perception of urban scenes in production environments. Architect and optimize complex machine learning systems for scalability, efficiency, and robustness, utilizing cloud-native technologies. Develop and implement training techniques such as data augmentation or model distillation pipelines to improve model robustness under diverse urban and environmental conditions. Explore and integrate novel multi-modal foundational AI models, including large Vision‑Language Models (VLMs), and develop strategies for their effective fine‑tuning and adaptation to specific domain challenges.

Drive innovation in model compression, quantization, and efficient inference techniques to optimize performance for both cloud and edge device deployments. Collaborate with cross-functional teams to define machine learning roadmaps, evaluate new technologies, and contribute to the overall technical strategy. Conduct research, and evaluate emerging deep learning techniques applicable to perception and intelligent mobility.

Design, build, implement and optimize multi‑stage computer vision pipelines that span segmentation, object detection, multimodal LLM/LVM extraction, machine‑readable code decoding, and multi‑source reconciliation. Train or fine‑tune detection models on custom medical supply datasets. Build and own dataset strategy including leveraging augmentation and synthetic data generation to improve training and testing datasets when data is scarce. Monitor and improve pipeline accuracy by instrumenting field‑level metrics, diagnosing failure modes, and systematically improving precision and recall through model iteration and preprocessing optimization.

Design persistence schemas and audit data models to ensure every data extraction is independently reviewable. Maintain and extend asynchronous Python backend services that provide pipeline results to downstream clinical workflows.

Design and implement algorithms for 3D point cloud processing, object detection, and segmentation. Enhance and optimize SLAM pipelines for real‑time mobile and static environments. Work on camera‑LiDAR calibration and multi‑sensor data alignment. Apply deep learning techniques to 2D, 3D, and spatial data. Integrate and optimize tools such as Open3D and 2D+3D inference models into existing systems. Optimize computer vision models and pipelines for NVIDIA GPUs.

Perform profiling and performance optimization to improve latency and accuracy. Collaborate with cross‑functional teams across Pakistan and Hong Kong to deliver product features. Participate in R&D to evaluate new computer vision and robotics techniques. Ensure robustness and reliability of computer vision systems in real‑world conditions. Support and guide junior engineers through reviews and technical discussions.

The Computer Vision Engineer will deliver hands‑on computer vision work and architect technical solutions for complex project requirements. They will lead the technical delivery of computer vision projects and provide expert guidance to multidisciplinary teams throughout the development lifecycle. The role includes contributing expert computer vision insight to bids and identifying opportunities to integrate advanced visual intelligence into customer solutions. The engineer will stay at the forefront of the field by mastering State‑of‑the‑Art developments and sharing best practices across the business unit.

They will represent the organization internally and externally as a subject matter expert in computer vision, partner with leadership to define the technical…